Research projects in Information Technology

This project focuses on developing a gait detection system leveraging computer vision techniques to recognize individuals in security footage, even when their faces and skin are obscured. Criminals often cover their faces to avoid recognition, making traditional facial recognition unreliable. By analyzing key features of an individual’s gait, such as walking style, speed, and direction, this system provides a robust alternative for visual attribution.

This project focuses on simulating the organic growth of criminal communication networks by leveraging techniques such as Reinforcement Learning and Graph Theory. The goal is to curate a synthetic dataset that models the evolving structure and dynamics of illegal networks, taking into account factors like social connections, communication patterns, and resource allocation. By using graph-based models, the project aims to create realistic representations of how criminal groups form, expand, and operate under various conditions.

In many branches of science (e.g., Artificial Intelligence, Engineering etc.), the modelling of the problem is done through the use of functions (e.g., f(x) = y). On a very high-level, we can think of Machine Learning as the problem of approximating function f from the pair of measurements (x,y), and Optimization as the problem of finding the value of input x that maximizes the output y given function f.

Overview

This project proposes a novel quantum-enhanced learning analytics framework for higher education, focusing on early identification of at-risk students and optimisation of intervention strategies using hybrid quantum-classical approaches. While current learning analytics systems rely on classical statistical and machine learning techniques, they often struggle to capture the complex, uncertain, and multi-dimensional nature of student learning behaviours.

Background and Motivation

Modern deep learning models have achieved remarkable success in computer vision and natural language processing. However, they typically produce overconfident predictions and lack reliable mechanisms to quantify uncertainty. This limitation becomes particularly problematic in high-stakes applications, such as healthcare diagnosis, autonomous systems, and scientific discovery.

Project Background & Motivation

Project Background and Motivation

The "inverse design" of molecules from analytical spectra (such as MS2, NMR, or IR) is a fundamental bottleneck in analytical chemistry, metabolomics, and drug discovery. While deep generative models have shown promise in proposing novel molecular structures, they typically require massive, cleanly labelled datasets to train effectively.

This PhD project focuses on the design and evaluation of hybrid quantum–classical algorithms for large-scale data analytics and optimisation problems.

The research will investigate how quantum computational techniques can be combined with classical systems to improve performance, scalability, and solution quality for tasks such as:

We are seeking a motivated PhD candidate to work on unsupervised music emotion tagging within the broader field of affective computing. The project aims to develop reproducible machine learning approaches for automatic emotion recognition in music, with stronger theoretical grounding, transparent model implementation, and rigorous validation.

Using the Project-1 hiPSC platform, this project builds AI pipelines to learn disease-relevant representations from cellular images, fused with multi-omics. Models will classify diagnosis and predict treatment response with strict donor-level splits, cross-regional external validation, and fairness audits (sex/ethnicity stratification). Interpretable AI (e.g., attribution maps, SHAP) will nominate mechanism-anchored biomarkers and candidate interventions. Tooling will be containerised and open to support reproducibility and clinical translation.